8. hausmann, leslie rm; schofield, janet ward; woods, rochelle l. sense of belonging as a predictor...

SENSE OF BELONGING AS A PREDICTOROF INTENTIONS TO PERSIST AMONG AFRICANAMERICAN AND WHITE FIRST-YEAR COLLEGESTUDENTS

Leslie R. M. Hausmann,*,† Janet Ward Schofield,** and Rochelle L.Woods***

................................................................................................................................................................................................

This study investigates the role of students’ sense of belonging to their university incollege student retention. Using individual growth curve modeling, we examined (a)whether sense of belonging predicts intentions to persist, and (b) the effects of anintervention designed to enhance students’ sense of belonging. African Americanand white first-year students completed surveys three times throughout theacademic year. Students were randomly assigned to a group that received anintervention to enhance students’ sense of belonging or to one of two controlgroups. Sense of belonging was found to predict intentions to persist, controlling forbackground variables and other predictors of persistence. Overall, sense ofbelonging and intentions to persist declined over the academic year. However, thedecline in sense of belonging was smaller for students in the intervention group.Implications for the development of college retention programs and for existingmodels of student persistence are discussed.

................................................................................................................................................................................................KEY WORDS: sense of belonging; persistence; intentions; intervention.

INTRODUCTION

Copious research over the last 30 years has identified numerousindicators of college student persistence. The most influential model ofstudent persistence (Tinto, 1987, 1993) identifies integration into the

*Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System,

University Drive C (151C-U), Pittsburgh, PA, USA.

**Department of Psychology, University of Pittsburgh, Pittsburgh, PA, USA.

***Office of Multicultural Initiatives, University of Michigan, Ann Arbor, MI, USA.

†Address correspondence to: Leslie R. M. Hausmann, Center for Health Equity Research

and Promotion, VA Pittsburgh Healthcare System, University Drive C (151C-U), Pittsburgh,

PA 15240, USA. E-mail: [email protected].

803

0361-0365/07/1100-0803/0 � 2007 Springer Science+Business Media, LLC

Research in Higher Education, Vol. 48, No. 7, November 2007 (� 2007)DOI: 10.1007/s11162-007-9052-9

social and intellectual fabric of the institution, as well as commitmentto the institution and to the goal of obtaining a college degree, as themost important predictors of student persistence (Donovan, 1984;Mayo, Murguia, and Padilla, 1995; Moores and Klas, 1989; Munro,1981; Pascarella, Duby, and Iverson, 1983; Pascarella and Terenzini,1980; Pascarella, Terenzini, and Wolfle, 1986). Other research hasfocused on additional factors including institutional fit (Bean, 1985;Cabrera, Castaneda, Nora, and Hengstler, 1992; Cabrera, Nora, andCastaneda, 1993), support from friends and family (Bank, Slavings, andBiddle, 1990; Mallinckrodt, 1988; Nora, 1987), financial need (Cabrera,Stampen, and Hansen, 1990; Hu and St. John, 2001; Mallette and Cab-rera, 1991; Nora, 1990; Nora, Cabrera, Hagedorn, and Pascarella,1996), racial climate (Cabrera, Nora, Terenzini, Pascarella, and Haged-orn, 1999; Eimers and Pike, 1997; Hurtado, 1992; Nora and Cabrera,1996), student involvement on campus (Berger and Milem, 1999; Milemand Berger, 1997), and institutional-level variables such as whether insti-tutions award 2-year or 4-year degrees (Strauss and Volkwein, 2004).The primary goal of this research is to call for a more systematic

study of one variable that has received sparse attention in existing stud-ies of student persistence: students’ sense of belonging to their college oruniversity, defined as the psychological sense that one is a valued mem-ber of the college community. This research has two specific objectives.The first is to examine the role of sense of belonging in predicting col-lege students’ intentions to persist, taking into account other variablesthat predict persistence. The second is to test the effects of an interven-tion designed to increase students’ sense of belonging.

Sense of Belonging in Models of Student Persistence

Failing to achieve an adequate sense of belonging can have impor-tant negative consequences (Baumeister and Leary, 1995; Durkheim,1951). Hence, it is appropriate that models of college student persis-tence have often included elements of belonging. The most widelyknown and studied model of student persistence is that of Tinto (1987,1993). Building upon the work of Spady (1971), Tinto theorized thatstudents’ integration into their social and academic college environ-ment predicts whether they are likely to remain enrolled in college. Inhis longitudinal model, students’ pre-college characteristics (e.g., socio-economic status, high school achievement, etc.) shape their initial levelof commitment to finishing college (goal commitment) and to complet-ing a degree at the college in which they are enrolled (institutionalcommitment). Initial commitment levels influence students’ social

804 HAUSMANN, SCHOFIELD AND WOODS

and academic experiences, which in turn affect social and academicintegration. Social and academic integration, along with initial goaland institutional commitment levels, determine subsequent levels ofgoal and institutional commitment, which ultimately determine stu-dents’ likelihood of departing from the institution before completingtheir degree. Tinto’s model also recognized that obligations outside ofthe college setting, such as work or family responsibilities, could alsoaffect levels of commitment.The feature of Tinto’s model that this paper highlights is related to

the mechanism by which it explained student departure - integrationinto the college setting. His model predicted that to the extent that stu-dents do not become integrated members of the college community theyare more likely to withdraw. Implicit in his theory was that a sense ofbelonging, as determined by social and academic integration, is a centralfeature of student persistence.Alternative models of student persistence share Tinto’s emphasis on

integration, and thus sense of belonging, in one form or another. Forinstance, Astin (1984) emphasized student involvement as essential topersistence. In Astin’s work, student involvement referred to actualstudent behaviors rather than to perceptions of social or academicintegration. Berger and Milem (1999; Milem and Berger, 1997) havefound that both student involvement behaviors and perceptions ofintegration play a role in persistence decisions. Specifically, they foundthat first-year students who reported more involvement behaviorsreported higher academic and social integration, as well as more insti-tutional commitment. Social integration, in turn, was associated withinstitutional commitment, intentions to enroll for a second year(Milem and Berger, 1997), and actual re-enrollment (Berger andMilem, 1999). The emphasis on student involvement and perceivedintegration, both of which are likely to be correlated with sense ofbelonging, is consistent with the idea that developing a sense ofbelonging is important to college persistence.Also, Bean’s (1985) student persistence model identified academic, so-

cial-psychological, and environmental factors likely to affect the sociali-zation of students, including variables closely related to sense ofbelonging. Specifically, the indicators of successful socialization includedin Bean’s model were institutional fit, college academic performance,and institutional commitment, all of which were hypothesized to affectpersistence. Institutional fit was described, in part, as the extent towhich students felt they ‘‘fit in’’ at the university, and was thus similarto sense of belonging.

SENSE OF BELONGING 805

Sense of Belonging in Student Persistence Research

As the previous discussion demonstrates, the concept of sense ofbelonging has been included in several models of student persistence inone form or another. However, sense of belonging is most often impliedas the result of social and academic integration, rather than specifiedand measured as an independent construct. In research examining vari-ous models of student persistence, sense of belonging is rarely, if ever,directly assessed so that its independent effects on persistence can bemeasured.In studies that do include at least one item that measures sense of

belonging, this item is often grouped with other items to form a mea-sure of a broader construct such as institutional fit (Bean, 1985; Cabreraet al., 1992) or institutional commitment (Cabrera et al., 1993; Noraand Cabrera, 1996; Strauss and Volkwein, 2004). Nora and Cabrera(1993) conducted a confirmatory factor analysis to investigate theappropriateness of combining sense of belonging with other measures toassess institutional commitment. Although the authors ultimately arguedfor a 2-factor structure of institutional commitment in which sense ofbelonging was combined with several other items, the factor structure inwhich sense of belonging was retained as a unique factor yielded a bet-ter fit to the data as measured by five out of six fit indices. Thus, thereis empirical justification for studying sense of belonging as a unique var-iable in student persistence research.There is also conceptual justification for examining sense of belonging

in studies of student persistence. Most persistence studies include mea-sures of social and/or academic integration, but the psychological sensethat one is an accepted member of one’s community, or sense of belong-ing, is distinct from one’s level of involvement with the community(Hurtado and Carter, 1997). Both students’ involvement and their sub-jective sense of belonging were important in early work on student per-sistence, in which students’ subjective sense of belonging was believed tomediate the relationships between student involvement and outcomessuch as student commitment and persistence (Spady, 1971). Hurtadoand Carter (1997) point out that subsequent work has almost com-pletely neglected the role of subjective sense of belonging. They take astep towards rectifying this oversight by studying the antecedents ofsense of belonging among a sample of Latino students. They also callfor future work to study the consequences of sense of belonging for out-comes such as student persistence.There is evidence supporting the idea that sense of belonging should

be considered in such research. For example, Zea, Reisen, Beil, and


Caplan (1997) assessed the collective self-esteem (Luhtanen andCrocker, 1992) students derived from their membership in the universitycommunity (i.e., how much students derived a positive identity fromtheir university). Zea et al. found that students who reported more col-lective self-esteem were less likely to report having thoughts about leav-ing the university. Although this study did not measure sense ofbelonging, per se, its results are consistent with the idea that having astrong sense of belonging is important for student persistence. Evenmore convincing support comes from Gaertner and Dovidio’s (2000)reanalysis of data from a study on racial climate at a large university(Snider and Dovidio, 1996) which found that students’ feelings ofbelonging to the university community mediated the positive relation-ship between satisfaction and intentions to finish their degree.We agree with Hurtado and Carter (1997) that sense of belonging has

been under-studied in student persistence research. Thus, in this studywe assess the correlates of sense of belonging in a sample of white andAfrican American freshmen. We also report the outcomes of an inter-vention designed specifically to increase students’ sense of belonging totheir university.

METHODOLOGY

Student Sample

The study was conducted at a large, public, mid-Atlantic, predomi-nately white (77% white, 8% African American, and 12% other race/ethnicity, 3% unknown race/ethnicity) university. A sample of full-time,first-year, non-transfer students were invited to participate in a three-wave survey during their first year of college. All first-year AfricanAmerican students (N = 254) and a random sample of 291 of theirwhite peers were invited to participate.The first survey was mailed during the second week of the fall semes-

ter. For Survey 1, 220 (76%) white and 145 (57%) African Americanstudents responded. All students who returned Survey 1 and who hadnot subsequently withdrawn from the university were invited to com-plete Surveys 2 and 3. The response rate for Survey 2, which was mailedduring the first week of the spring semester, was 94% for both whitesand African Americans. Response rates for the third survey, which wasmailed during week 11 of 17 of the spring semester, were also high(96% for whites, 89% for African Americans). Median response timesfor Surveys 1, 2, and 3 were 14, 7 and 11 days, respectively.


Those who returned at least one survey, and hence were included inthe analyses, were between 16 and 21 years of age (Mean = 18,SD = 0.49). The majority of the sample lived on campus (83%), withthe remainder living in off-campus housing (1%) or with family (16%).Approximately 60% (219/365) of the total sample was female. Thewhite and African American samples were 55% and 68% female,respectively. Women were slightly over-represented in our sample com-pared with the university population, in which 51% of white studentsand 59% of African American students are female. Students were paidfor completing each survey and provided written informed consentbefore participating in the study.

Procedure

A longitudinal experimental design was used. Participants completeda survey containing measures of financial difficulties, social and aca-demic integration, peer and parental support, sense of belonging, insti-tutional commitment, and intentions to persist at the beginning of theirfirst semester and at the beginning and end of their second semester.Upon returning the first survey, respondents were randomly assigned toan enhanced sense of belonging group or one of two control groupswith the constraint that white and African American students were dis-tributed equally across each group.A multi-faceted approach was designed to increase sense of belonging

in students in the enhanced sense of belonging group. These studentsreceived several written communications from university administrators(e.g., the Provost and/or Vice-Provost for Student Affairs) emphasizingthat they were valued members of the university community and thattheir responses to the surveys (in aggregate form) would be used to helpimprove campus life for all students. These students also received smallgifts for daily use (e.g., ID holders, magnets, decals, etc.) that displayedthe university’s name, logo, and colors. The purpose of these gifts wasto surround students with items that emphasized their connection totheir university.Students in both control groups were asked to complete the same sur-

veys but did not receive the communications and logo-bearing giftsdesigned to affect students’ sense of belonging. Specifically, all commu-nication with these students came from a professor in the Psychologydepartment rather than from university administrators. Furthermore,students’ membership in the campus community was not mentioned inthese communications. We thought it was possible that receiving giftsduring one’s freshman year might be sufficient to affect students’ sense


of belonging, regardless of whether the gifts displayed university insig-nia. Therefore, students in the gift control group received paraphernaliafrom the psychology professor identical to that received by students inthe enhanced sense of belonging group except that the gifts for thisgroup did not contain university insignia, name, or colors. In the no-giftcontrol group, students did not receive any gifts or additional communi-cations, thus providing data from respondents who did not have anyexperiences related to their participation in the study other than comple-tion of the surveys.

Variables Studied

Our goal was to examine the unique role of sense of belonging in stu-dent persistence, while taking into account other factors that have beenshown to predict student persistence. Because the key determinants ofstudent persistence in Tinto’s model are (1) pre-college characteristics,(2) social and academic integration, and (3) student commitment, weincluded measures of these variables in this study. We also included twoadditional variables, peer and parental support, that have been shownto be reliable predictors of student persistence (Bean, 1980; Cabreraet al., 1992, 1993, 1999; Eimers and Pike, 1997; Nora et al., 1996).Table 1 describes the variables measured in this study, each of which isdiscussed below, as well as provides information on their reliability,means, and standard deviations.

Student Background Variables

We included four student background variables: race, gender, finan-cial difficulties, and SAT scores (see Table 1). Students’ perceived finan-cial difficulty was assessed because students’ attitudes regarding thefinancing of their college education have been shown to relate to theirpersistence behavior (Cabrera et al., 1992; Mallette and Cabrera, 1991).Students’ SAT scores were included as a measure of academic back-ground.

Social and Academic Integration

We used Pascarella and Terenzini’s (1980) measures of social and aca-demic integration. These included two subscales (peer-group interac-tions; interactions with faculty) designed to assess social integration andtwo subscales (perceived faculty concern for student development andteaching; academic and intellectual development) designed to assess


TABLE

1.DescriptionofStudyVariables

Variable

Description

Alpha

Mean(SD)

Tim

e1

Tim

e2

Tim

e3

Tim

e1

Tim

e2

Tim

e3

Studentbackgroundcharacteristics

Race

White=

1

AfricanAmerican=

)1

Gender

Male

=1

Fem

ale

=)1

Financialdiffi

culty

(FIN

DIF

F)

Averageresponse

tothefollowingtw

o

item

s:Considering

the

financialaid

you’vereceived

andthemoney

youand

yourfamilyhave,

how

much

diffi

culty

haveyouhadso

farin

payingforyour

schoolexpenses?

(no

diffi

culty,some

diffi

culty,oragreatdealofdiffi

culty)

Doyouhaveanyconcern

aboutyour

future

abilityto

finance

yourcollege

education?(no,someconcern,orma-

jorconcern).

Higher

values

indicate

greaterfinancialdiffi

culty.

0.78

0.83

0.83

1.88(0.61)

1.93(0.64)

1.97(0.66)

SAT

comprehensive

(SAT)

Sum

ofSAT

verbaland

quantitative

scores.

Obtained

from

university

records.

1187(141)


TABLE

1.(Continued)

Variable

Description

Alpha

Mean(SD)

Tim

e1

Tim

e2

Tim

e3

Tim

e1

Tim

e2

Tim

e3

Integrationvariables

Interactions

withpeers

(PEERIN

T)

Averagescore

ontheinteractionswith

peers

subscale

inPascarellaandTer-

enzini(1980)

0.78

0.79

0.82

3.54(0.74)

3.73(0.72)

3.75(0.75)

Interactions

withfaculty

(INTFAC)

Averagescore

ontheinteractionswith

faculty

subscale

inPascarella

and

Terenzini(1980)

0.79

0.83

0.85

3.11(0.69)

3.25(0.76)

3.28(0.78)

Academ

ic

integration

(ACADIN

T)

Averagescore

onthefacultyconcern

and

academ

icdevelopmentsubscales

inPascarellaandTerenzini(1980)

0.76

0.80

0.82

3.72(0.52)

3.70(0.58)

3.67(0.57)

Support

variables

Peersupport

(PEERSUP)

My

close

friends

encourage

me

to

continueattending<

nameofinstitu-

tion>

.(From

Cabrera

etal.,1992)

3.98(1.02)

4.01(1.04)

4.01(1.05)

Parental

support

(PARSUP)

Averagescore

ontw

oitem

s:Myfam-

ilyapproves

ofmyattending<

name

ofinstitution>

.Myfamilyencourages

meto

continueattending<

nameof

institution>

.(From

Cabrera

etal.,

1992)

0.78

0.73

0.69

4.66(0.64)

4.51(0.73)

4.51(0.69)


TABLE

1.(Continued)

Variable

Description

Alpha

Mean(SD)

Tim

e1

Tim

e2

Tim

e3

Tim

e1

Tim

e2

Tim

e3

Sense

ofbelonging

(BELONG)

Averagescore

on

sense

ofbelonging

subscale

inBollen

and

Hoyle

(1990).

Item

sincluded:

Ifeel

asense

of

belongingto

<nameofinstitution>

.

Iam

happyto

beat<

nameofinsti-

tution>

.Iseemyself

aspart

ofthe

<nameofinstitution>

community.

0.89

0.93

0.93

4.00(0.87)

3.99(0.91)

3.93(0.92)

Institutional

commitment

(COMMIT

)

Average

score

on

two

item

sfrom

institutionaland

goalcommitments

subscalefrom

PascarellaandTerenzini

(1980):

Itis

notim

portantformeto

graduate

from

this

university.Iam

confidentImadetherightdecisionto

attend<

nameofinstitution>

.

0.51

0.58

0.50

4.34(0.90)

4.14(0.96)

3.95(1.04)

Intentionsto

persist

(INTENT)

Iintend

tocomplete

my

degree

at

<nameofinstitution>

.

4.49(0.93)

4.31(1.04)

4.24(1.14)

Note:Allintegrationvariables,

support

variables,

sense

ofbelonging,institutionalcommitment,andintentionsto

persist

weremeasuredusing

5-pointLikertresponse

scalesonwhich1=

strongly

disagreeand5=

strongly

agree.Negativelyworded

item

swerereverse-scoredpriorto

being

combined

withother

item

s.Higher

values

indicate

more

favorable

responsesforallmeasures.


academic integration. We conducted a confirmatory factor analysis inwhich different subscale combinations were examined.1 A factor struc-ture in which the peer-group interactions and interactions with facultysubscales remained separate, and the faculty concern and academic andintellectual development subscales were combined, yielded the best fit tothe data. We thus included average scores for the peer-group interac-tions and interactions with faculty subscales as separate predictors inthe analyses. Faculty concern and academic and intellectual develop-ment subscale items were averaged into a single score representing aca-demic integration (see Table 1).

Peer and Parental Support

We utilized items measuring peer and parental support from Cabreraet al. (1992). More details regarding these items are provided in Table 1.

Sense of Belonging

Bollen and Hoyle (1990) proposed that perceived cohesion, or ‘‘theextent to which group members feel ‘stuck to,’ or part of, particularsocial groups’’ (p. 482), consists of individuals’ sense of belonging to agroup as well as their sense of morale regarding group membership.Their conceptualization of sense of belonging as feeling part of a socialgroup is very similar to Spady’s (1971) conceptualization of students’subjective sense of integration into the college community. We thus usedBollen and Hoyle’s sense of belonging subscale in this study (seeTable 1).

Institutional Commitment

Institutional commitment was measured using the average response totwo items from Pascarella and Terenzini’s (1980) institutional and goalcommitments subscale (see Table 1).2, 3

Intentions to Persist

Intentions to persist were assessed with a single item (see Table 1).Although intentions to persist and actual persistence are not equivalent,the use of intentions to persist as an outcome in persistence research issubstantiated by studies that show a strong association between inten-tions to persist and actual persistence (e.g., Bean, 1980; Cabrera et al.,1992, 1993; Pascarella et al., 1983).


Analytic Procedures

Multilevel Model for Change

Data were analyzed using the multilevel model for change (MMC)technique (Singer and Willett, 2003), also known as individual growthmodeling. A special case of hierarchical linear modeling, this techniqueinvolves analyzing data at multiple levels. Rather than grouping thedata according to a between subject characteristic, MMC involvesgrouping data from the same individuals across time. In MMC analyses,all variables are measured at all time points, thus allowing one to exam-ine: (a) whether variables change over time, and (b) what factors areassociated with the change variables undergo over time. This procedureinvolves estimating the starting value (i.e., initial status) and changeover time (i.e., rate of change) for the variables of interest.

Data Preparation

Multilevel model for change requires the creation of a person–perioddataset in which each line contains data from one person at one timepoint. For example, a student who returned all three surveys wouldhave three separate lines of data in the person–period dataset, eachidentified by that student’s identification number and the time at whichthe student returned that particular survey to the researchers. The initialperson period data set contained 1042 lines of data. Sixteen lines ofdata were deleted due to missing values.Variables with a slight negative skew were transformed prior to analy-

ses. An inverse transformation on the reverse-scored values best normal-ized the distributions of the following variables: peer support, parentalsupport, institutional commitment, and intentions to persist. Peer-groupinteractions and sense of belonging were best normalized using a log(base 10) transformation on the reverse-scored values. The transformedvalues of peer group interactions and sense of belonging were multipliedby )1 to re-establish the original direction of the scale (i.e., higher val-ues indicate more positive responses). Interactions with faculty, aca-demic integration, financial difficulty, and SAT scores did not requiredata transformations. All continuous variables were mean-deviated priorto the analyses. Parameter estimates for a given predictor reflect thevalue of that predictor when all other predictors are at their mean value(i.e., zero). All reported parameter estimates are unstandardized andreflect the transformations performed on the data.


The metric of time used in this study is the number of weeks (i.e.,days/7) that elapsed between the start of the academic year (time = 0)and the return of each completed survey. The reported parameter esti-mates for rate of change represent the magnitude of change in a vari-able over a period of 1 week.

Analytic Strategy

The analyses proceeded in several stages. First, we determined whe-ther there was sufficient variation in several variables to justify exploringpredictors associated with that variation. This was achieved by estimat-ing the unconditional means model for peer group interactions, interac-tions with faculty, academic integration, peer support, parental support,sense of belonging, institutional commitment, and intentions to persist.The unconditional means model also provides a baseline model forcomparison with subsequent models. For this and all other models, the)2LL deviance statistic was used to assess goodness-of-fit.Next, we determined whether a model that includes time as a predic-

tor provides a better fit to the data than a model that does not includetime. This was achieved by estimating the unconditional growth model,which is the unconditional means model plus the effect of time, for thesame set of variables. Chi-square tests comparing goodness-of-fit statis-tics for the unconditional means and unconditional growth models indi-cate whether consideration of time improves fit. The informationproduced by the unconditional growth model also indicates whetherthere is sufficient variation in rate of change across participants toexplore predictors associated with that variance.After that we estimated models in which student background charac-

teristics (i.e., race, gender, financial difficulty, and SAT scores) were in-cluded as predictors of the Time 1 value (initial status) and the changebetween Times 1, 2 and 3 (rate of change) of the variables found to im-pact persistence in previous research (i.e., integration variables: peer-group interactions, interactions with faculty, academic integration; sup-port variables: peer support and parental support). Separate models foreach integration and support variable were estimated and the resultingdeviance statistics were compared to the deviance statistics for theunconditional growth model for each variable. This indicated whethermodels containing student background variables fit the data better thanmodels that contained only time.We then estimated a model in which all student background, integra-

tion, and support variables were included as predictors of the initial sta-tus and rate of change of sense of belonging. This model also tested the


effect of the intervention on the rate of change of sense of belonging.Because the intervention did not begin until after the first survey wascompleted, the effect of the intervention on initial status was notincluded in the model. Effects of the intervention were tested by threepairs of orthogonal contrast codes (Judd and McClelland, 1989, seeTable 2). Each pair of contrast codes was included in a separate estima-tion of the model. Because orthogonal contrast codes were used, the fitof the model and parameter estimates of the other variables in the mod-el were identical regardless of which pair of contrast codes was includedin the model.Because institutional commitment is so often emphasized as having

a crucial role in influencing persistence, the above analysis wasrepeated using institutional commitment as the outcome variable andincluding sense of belonging as an additional predictor for both initialstatus and rate of change. The analysis was also repeated with inten-tions to persist as the outcome, with sense of belonging and institu-tional commitment as additional predictors of initial status and rateof change.Finally, to determine whether results varied by race, interactions

between race and all predictors were added to each model. Whether thisresulted in better-fitting models than those in which no race interactionswere included was assessed.

TABLE 2. Contrast-coded (CC) Effects of the Intervention

Experimental group

Comparison

Enhanced

sense of

belonging (ESB)

Gift control

(GC)

No-gift

control (NGC)

CC1A 2 )1 )1 ESB versus GC and

NGC combined

CC1B 0 1 )1 GC versus NGC

CC2A 1 )1 0 ESB versus GC

CC2B 1 1 )2 ESB and GC combined

versus NGC

CC3A 1 0 )1 ESB versus NGC

CC3B )1 2 )1 ESB and NGC combined

versus GCa

aCC3B is not of theoretical interest but is the necessary complement to CC3A.


RESULTS

Unconditional Means Models (UMM)

Estimates of the UMMs are reported in Table 3. The parameter esti-mates for the intercept in initial status indicate whether the grand meanof a given outcome varies significantly from zero. None of them did,which was expected because all of the outcome variables were mean-deviated prior to analyses (i.e., the mean = 0).The most informative elements of Table 3 are the variance compo-

nents. The within-person variance component of the UMM indicateswhether there is significant variability within individual students’responses. The initial status variance component indicates whether thereis significant between-person variability in students’ average responses.The within-person and initial status variance components are statisti-cally significant for all outcome variables. Predictors were added to sub-sequent models with the goal of explaining some of this variance.

Unconditional Growth Models (UGM)

Estimates of the UGMs are reported in Table 4. The UGM is identi-cal to the UMM except that it contains a measure of time as a predic-tor. Comparing the goodness-of-fit statistics for the UGM to those ofthe UMM indicates whether including time in the model resulted in asignificantly better fit with the data. The deviance statistics for theUGMs and UMMs were compared using chi-square analyses. Thedegrees of freedom for these tests are the difference in the number ofparameters estimated in each of the two models, including fixed effectsand variance components. Including time as a predictor resulted in abetter-fitting model for all variables except peer support (see Table 4).This indicates that examining all variables except peer support longitu-dinally yields more information than examining them cross-sectionally.The UGM includes two fixed effects: the initial status intercept and

the rate of change intercept. The initial status intercept of the UGMrepresents the average starting value of the outcome (i.e., average Time1 response), and the parameter estimates for the initial status interceptsindicate whether the average Time 1 value of each outcome differed sig-nificantly from zero. Because all variables were mean-deviated, an initialstatus intercept that is not different from zero indicates that the meanvalue of the variable at Time 1 is not different from the mean value ofthe variable collapsed across all time points. The initial status interceptparameter estimates indicate that the starting values of peer-group


TABLE

3.UnconditionalMeansModels

PEERIN

TIN

TFAC

ACADIN

TPEERSUP

PARSUP

BELONG

COMMIT

INTENT

Fixed

effects

Initialstatus

Intercept

)0.0004

0.0041

)0.0031

)0.0016

)0.0017

)0.0005

)0.0019

)0.0006

Variance

components

Within-person

0.0075*

0.2912*

0.1307*

0.0568*

0.0305*

0.0145*

0.0424*

0.0455*

Initialstatus

0.0121*

0.2627*

0.1793*

0.0387*

0.0335*

0.0208*

0.0375*

0.0429*

Goodness-of-fit

Deviance

()2LL)

)1493.61

2103.10

1395.78

355.27

)160.67

)848.31

121.90

209.30

Note:

*p<

0.001;Parameter

estimates

are

unstandardized.PEERIN

T=

peer-group

interactions,

INTFAC

=interactions

with

faculty,

ACADIN

T=

academ

icintegration,

PEERSUP=

peer

support,

PARSUP=

parental

support,

BELONG

=sense

of

belonging,

COMMIT

=institutionalcommitment,IN

TENT=

intentionsto

complete

degreeatcurrentuniversity.


interactions and interactions with faculty were significantly lower thanthe mean of those outcomes across all time points. The starting valuesfor academic integration, peer support, and sense of belonging did notdiffer significantly from the overall mean of those outcomes. Parentalsupport, institutional commitment, and intentions to persist all hadstarting values that were significantly higher than the overall mean ofthose outcomes.The rate of change intercept in the UGM represents the average rate

of change across all participants for the outcome variable, not control-ling for the effects of any variables. A significant rate of change inter-cept indicates that, on average, the value of that outcome changes overtime. As one might expect, peer-group interactions and interactions withfaculty increased over time. In contrast, parental support, institutionalcommitment, and intentions to persist decreased over the duration ofthe study. There was also a marginally significant decrease over time foracademic integration and no average change over time for peer supportor sense of belonging.Although the fixed effects in the UGM are interesting, the variance

components are more informative for subsequent model testing. Thepresence of time in the model alters the meaning of the variance compo-nents. The within-person variance component now represents the vari-ability of individuals’ responses around their unique linear changetrajectories rather than their unique mean responses. The initial statusvariance component now represents between-person variability in initialstatus of the outcome (i.e., Time 1 response value), rather than between-person variability in average response. The addition of time to the mod-el results in two additional variance components: covariance and rate ofchange. The covariance component represents the extent to which initialstatus co-varies with rate of change. The rate of change variance com-ponent represents between-person variability in rate of change, or theextent to which people deviate from the average rate of change for agiven outcome variable.The most informative variance components are those for within-per-

son, initial status, and rate of change. The within-person variance com-ponents in Table 4 indicate that there is significant variation ofindividuals’ scores around individuals’ unique change trajectories for alloutcome variables. This raises the possibility of identifying predictors toaccount for some of this variation. A similar pattern of results occurredfor variance in initial status. That is, there was significant between-per-son variability in starting values for all outcomes. There was significantbetween-person variability in rate of change for peer-group interactions,interactions with faculty, sense of belonging, and intentions to persist.


TABLE

4.UnconditionalGrowth

Models

PEERIN

TIN

TFAC

ACADIN

TPEERSUP

PARSUP

BELONG

COMMIT

INTENT

Fixed

effects

Initialstatus

Intercept

)0.0283***

)0.1273***0.0332

)0.0084

0.0484***

0.0094

0.0772***

0.0502**

Rate

ofchange

Intercept

0.0016***

0.0074***

)0.0021

†0.0004

)0.0029***

)0.0005

)0.0045***

)0.0029***

Variance

components

Within-person

0.0054***

0.2380***

0.1176***

0.0514***

0.0258***

0.0114***

0.0343***

0.0309***

Initialstatus

0.0131***

0.2127***

0.1520***

0.0440***

0.0307***

0.0244***

0.0425***

0.0525***

Covariance

)0.0001

†)0.0002

0.0003

)0.0004

)0.00002

)0.0002*

)0.0003

)0.0007***

Rate

ofchange0.00001***

0.0002**

0.0001

†0.00003

†0.00002

†0.00002***0.00002

†0.0001***

Goodness-of-fit

Deviance

()2LL)

)1558.44

2052.77

1379.43

351.08

)204.17

)867.34

55.49

146.33

Chi-square

deviance

comparisonsbetweenunconditionalmeansandgrowth

models

Chi-square

64.83***

50.33***

16.34***

4.19

43.50***

19.03***

66.41***

62.97***

Note:

†p<

0.1,*p<

0.05,**p<

0.01,***p<

0.001;Parameter

estimatesare

unstandardized.df=

3forallChi-square

comparisons;

PEER-

INT=

peer-groupinteractions,

INTFAC

=interactionswithfaculty,ACADIN

T=

academ

icintegration,PEERSUP=

peersupport,PAR-

SUP=

parentalsupport,BELONG

=sense

ofbelonging,COMMIT

=institutionalcommitment,IN

TENT=

intentionsto

complete

degreeat

currentuniversity.


There was only marginally significant variability in rate of change foracademic integration, peer support, parental support, and institutionalcommitment. Low variability in rate of change foreshadows a difficultyin identifying predictors that are associated with rate of change, giventhat there is little variance to be explained to begin with.

Integration and Support Variables

Table 5 displays models in which the student background variables ofrace, gender, SAT, and financial difficulty were used as predictors of theinitial status and rate of change for peer-group interactions, interactionswith faculty, academic integration, peer support, and parental support.Deviance statistics of each model containing the student backgroundpredictors and the corresponding UGMs were compared. As reported inthe last row of Table 5, including the student background predictorsresulted in a better-fitting model for peer-group interactions, interac-tions with faculty, academic integration, peer support, and parentalsupport.Examining the fixed effects for each model provides a better under-

standing of how each student background characteristic relates to theinitial status and rate of change of each outcome variable.4 For peer-group interactions, no background characteristics were significantlyrelated to the initial starting value. That is, students reported the samenumber of peer-group interactions at the beginning of the academicyear, regardless of race, gender, SAT score, or financial difficulty. Theincrease in peer-group interactions over time (i.e., the rate of changeintercept) remains significant after controlling for the student back-ground characteristics. The only additional significant predictor of rateof change is SAT scores. Students with higher SAT scores experienced amore rapid increase in their peer-group interactions than students withlower SAT scores.For interactions with faculty, the only significant effect of student

background variables was a negative effect of SAT score on initial sta-tus. Students with higher SAT scores reported fewer interactions withfaculty members at the beginning of the academic year than those withlower SAT scores. Students reported an increase in interactions withfaculty members over the course of the academic year, and this increasedid not depend on students’ race, gender, SAT score, or financial diffi-culty.For academic integration, both gender and SAT scores were associ-

ated with initial status. Males reported less academic integration thanfemales at the beginning of the academic year. Students with higher


TABLE

5.ModelsEstim

atingIntegrationandSupport

Variables,Sense

ofBelonging,InstitutionalCommitment,

andIntentionsto

Persist

PEERIN

TIN

TFAC

ACADIN

TPEERSUP

PARSUP

BELONG

COMMIT

INTENT

Fixed

effects

Initialstatus

Intercept

)0.0334***

)0.1283***

0.0141

)0.0238

0.0336*

0.0195*

0.0596***

0.0261

†

RACE

0.0107

)0.0024

0.0070

0.0408*

0.0559***

0.0104

0.0167

)0.0166

GENDER

)0.0098

0.0130

)0.0725*

)0.0255

)0.0070

0.0122

)0.0155

0.0214

SAT

)0.0021

)0.9852**

0.4885*

)0.1027

)0.0927

)0.0242

0.1227

0.0899

FIN

DIF

F)0.0134

)0.0273

)0.0432

)0.0231

)0.0183

0.0049

0.0101

)0.0235

PEERIN

T0.4853***

0.2349*

0.1232

INTFAC

0.0335**

)0.0089

0.0042

ACADIN

T0.0143

0.0487

†)0.0254

PEERSUP

0.0874***

0.0965*

0.0436

PARSUP

0.1781***

0.3101***

0.0653

BELONG

0.2915***

0.2961***

COMMIT

0.3602***

Rate

ofchange

Intercept

0.0016***

0.0081***

)0.0019

0.0003

)0.0027***

)0.0011**

)0.0037***

)0.0011

†

RACE

)0.0001

)0.0011

0.0003

0.0004

)0.0005

0.00003

0.0003

0.0010

GENDER

)0.0004

0.0010

)0.0001

)0.0006

)0.0003

)0.0004

0.0009

)0.00001

SAT

0.0062**

0.0100

0.0124

0.0011

0.0054

)0.0009

)0.0006

0.0010

FIN

DIF

F)0.0004

)0.0030

)0.0029

0.0002

)0.0006

)0.0003

0.00002

0.0004

CC1A

0.0005**

0.0001

0.0004

CC1B

0.0003

)0.00003

)0.0006

CC2A

0.0005

†0.0002

0.0008

†

CC2B

0.0004*

0.0001

)0.0001

CC3A

0.0009**

0.0002

0.0003

CC3B

0.0001

0.0001

0.0005

†


Table

5.(Continued)

PEERIN

TIN

TFAC

ACADIN

TPEERSUP

PARSUP

BELONG

COMMIT

INTENT

PEERIN

T)0.0035

0.0022

)0.0053

INTFAC

)0.0002

0.0003

)0.0001

ACADIN

T0.0022**

0.0004

0.0015

PEERSUP

)0.0008

0.0001

)0.0017

PARSUP

)0.0038*

)0.0019

0.0076*

BELONG

)0.0024

0.0064

COMMIT

)0.0041

Variance

components Within-person

0.0054***

0.2370***

0.1167***

0.0514***

0.0257***

0.0111***

0.0321***

0.0271***

Initialstatus

0.0128***

0.1989***

0.1428***

0.0419***

0.0283***

0.0098***

0.0188***

0.0293***

Covariance

)0.0001*

)0.0002

0.0001

)0.0004

)0.00002

)0.0001

)0.0003

)0.0008***

Rate

ofchange

0.00001***

0.00023**

0.00006

0.00003

0.00002�

0.00001

0.00001

0.00005***

Goodness-of-fit

Deviance

()2LL)

)1591.75

2032.61

1339.32

332.63

)230.03

)1261.31

)279.48

)237.32

Chi-square

deviance

comparisonswithunconditionalgrowth

models

df

88

88

820

22

24

Chi-square

33.31***

20.15**

40.11***

18.46*

25.87**

393.98***

334.97***

383.65***

Note:

†p<

0.1,*p<

0.05,**p<

0.01,***p<

0.001;Parameter

estimatesare

unstandardized.PEERIN

T=

peer-group

interactions,

INT-

FAC

=interactions

with

faculty,

ACADIN

T=

academ

icintegration,

PEERSUP=

peer

support,

PARSUP=

parental

support,

BE-

LONG

=sense

of

belonging,COMMIT

=institutionalcommitment,

INTENT=

intentions

tocomplete

degree

at

current

university,

FIN

DIF

F=

financialdiffi

culty.See

Table

2fordescriptionofcontrast-coded

effects.ForBELONG,COMMIT

,andIN

TENT,only

onepairof

contrast

codes

wasincluded

inthemodel

atatime.


SAT scores also initially reported more academic integration than thosewith lower SAT scores. There was no change in academic integrationover time, nor were any student background variables associated withchanges in academic integration.For peer support, the only significant effect for either initial status or

rate of change was the effect of race on initial status. White studentsreported more peer support than African American students at thebeginning of the academic year.For parental support, there was a significant effect of race on initial

status. White students reported more parental support at the beginningof the academic year than African American students. There was also asignificant decline in parental support over time, but this was not associ-ated with students’ race, gender, SAT score, or financial difficulty.

Sense of Belonging

The model in which student background, integration, support vari-ables, and contrast-coded effects of the intervention were used to predictsense of belonging is displayed in Table 5. Including this set of predic-tors resulted in a better-fitting model compared to the UGM.Starting values of sense of belonging were not related to students’

race, gender, SAT scores, or financial difficulty. However, greater senseof belonging at the start of the academic year was associated with peer-group interactions, interactions with faculty, peer support, and parentalsupport, but not academic integration.5

On average, students reported a small but statistically significant de-cline in sense of belonging over the course of the academic year, asshown in Table 1. This decline was not associated with any studentbackground characteristics. It was, however, associated with academicintegration and parental support. Students who reported more academicintegration experienced an increase in sense of belonging over time,whereas those with less academic integration experienced a decrease insense of belonging. Having more parental support, however, was associ-ated with a faster decline in sense of belonging over time.The effect of the intervention designed to increase students’ sense of

belonging was tested by three pairs of contrast codes (see Tables 2 and5). The mean values of sense of belonging for each group of partici-pants at each time point are shown in Table 6. Results for the first pairof contrast codes indicated that students in the enhanced sense ofbelonging group experienced a less rapid decline in sense of belongingover time compared to both of the control groups combined (CC1A),and that sense of belonging declined equally over time for students in


each of the control groups (CC1B). Results for the second set of con-trast codes indicated that there was a marginally significant trend forstudents in the enhanced sense of belonging group to experience a less-rapid decline in sense of belonging over time than those who were in thegift control group (CC2A), and that students who received gifts of anykind (containing university insignia or not) reported a less rapid declinein their sense of belonging over time than those who did not receivegifts (CC2B). Results for the final set of contrast codes indicated thatstudents in the enhanced sense of belonging group experienced a lessrapid decline in sense of belonging compared to students in the no-giftcontrol group (CC3A).6

Institutional Commitment

A model containing student background, integration, and supportvariables, sense of belonging, and the effects of the intervention as pre-dictors of institutional commitment is reported in Table 5. This modelfit the data better than the UGM.The fixed effects indicated that no student background characteristics

were associated with the initial status of institutional commitment.However, the initial status of institutional commitment was positivelyassociated with peer-group interactions, peer support, parental support,and sense of belonging. The association between initial status of institu-tional commitment and academic integration was marginally significant,and there was no association between institutional commitment andinteractions with faculty.Institutional commitment declined over time. Interestingly, the decline

was not associated with any of the predictors (see Table 5). Studentstend to report less commitment to the university over their first year ofcollege, regardless of background, integration experiences during college,

TABLE 6. Means (and Standard Deviations) of Sense of Belonging and Intentions

to Persist for the Intervention and Control Groups at Times 1, 2, and 3

Outcome Participant group Time 1 Time 2 Time 3

Sense

of belonging

Enhanced sense of belonging 4.05 (0.89) 4.13 (0.87) 4.04 (0.92)

Gift control 4.02 (0.85) 3.96 (0.84) 4.02 (0.89)

No gift control 3.94 (0.89) 3.87 (1.01) 3.75 (0.92)

Intentions

to persist

Enhanced sense of belonging 4.54 (0.96) 4.39 (1.04) 4.31 (1.14)

Gift control 4.47 (0.96) 4.31 (0.99) 4.23 (1.13)

No gift control 4.47 (0.87) 4.25 (1.09) 4.18 (1.14)


or support from peers or parents. The intervention also did not preventthis general decline in commitment.

Intentions to Persist

Finally, a model including all student background, integration, andsupport values, the effects of the intervention, sense of belonging, andinstitutional commitment as predictors of initial status and rate ofchange of intentions to persist was estimated (see Table 5). The modelincluding these predictors fit the data better than the UGM.Examining the fixed effects reveals that the intentions to persist stu-

dents report at the beginning of the academic year are not related tostudent background characteristics, integration experiences, or supportfrom friends or parents. However, both sense of belonging and institu-tional commitment are positively associated with initial status of inten-tions to persist. That is, students who report a greater sense ofbelonging or more institutional commitment at any time point also re-port stronger intentions to persist at the beginning of the academic year.There is a marginally significant trend for students to report weaker

intentions to persist over time. It is striking that very few predictors areassociated with this decline in intentions. Parental support is the onlypredictor that is significantly related to change in intentions over time,such that students who report having more parental support experiencean increase, rather than decrease, in intentions to persist. In general, theintervention did not unambiguously affect changes in intentions to per-sist. However, there was a marginally significant trend for students inthe enhanced sense of belonging group to experience a less-rapid declinein intentions compared to those in the gift control group.

Interactions with Race

To explore whether the results in Table 5 were consistent acrossAfrican American and white students, interactions between each predic-tor and race were added to each model, and each model was re-esti-mated. Deviance statistics for the new models and the models excludingthe race interactions were compared. Including the race interactionsimproved the fit only for the models predicting academic integrationand peer support (see Table 7). Improvement of the peer-group interac-tions model was marginally significant, as well.Based on the generally weak improvement in overall fit with the data

provided by including the race interactions, it is not surprising thatthere were few statistically significant fixed effects of the interactions


TABLE

7.Chi-square

Tests

of)2LLDeviance

StatisticsComparingModelsin

Table

5withModelsContainingAllRace*

PredictorInteractionsforBoth

InitialStatusandRate

ofChangePredictors

PEERIN

TIN

TFAC

ACADIN

TPEERSUP

PARSUP

BELONG

COMMIT

INTENT

Goodness-of-fitofmodelscontainingrace

interactions

Deviance

()2LL)

)1604.22

2026.32

1326.14

318.71

)233.91

)1280.68

)294.38

)266.38

Chi-square

deviance

comparisonswithunconditionalgrowth

models

df

66

66

620

20

22

Chi-square

12.48

†6.29

13.18*

13.92*

3.89

19.37

14.90

29.07

Note:

†p<

0.1,*p<

0.05,PEERIN

T=

peer-group

interactions,

INTFAC

=interactionswith

faculty,ACADIN

T=

academ

icintegration,

PEERSUP=

peer

support,

PARSUP=

parental

support,

BELONG

=sense

of

belonging,

COMMIT

=institutional

commitment,

INTENT=

intentionsto

complete

degreeatcurrentuniversity.


with race.7 For peer-group interactions, interactions with faculty,parental support, institutional commitment, and intentions to persist,there were no significant race interactions. For academic integration, theinteraction between gender and race was significantly related to initialstatus (parameter estimate = )0.091, p<0.01). This interaction sug-gested that white males reported less academic integration at the begin-ning of the academic year than did white females, but this pattern wasreversed and less pronounced for African American males and females.For peer support, there were significant interactions between gender andrace for both initial status (parameter estimate = )0.0602, p<0.001)and rate of change (parameter estimate = 0.0018, p<0.05). Whitemales reported less peer support at the start of the academic year thandid white females, but this pattern was reversed for African Americanstudents. Peer support increased over time for white males butdecreased over time for white females. This pattern was reversed andmore pronounced for African American students.The remaining three race interactions occurred in the sense of belong-

ing model. There was an interaction between race and parental supportfor both initial status (parameter estimate = )0.114, p<0.01) and rateof change (parameter estimate = 0.0043, p<0.01) of sense of belong-ing. Having more parental support was associated with greater sense ofbelonging at the beginning of the academic year for students of bothraces. This pattern was more pronounced, however, for African Ameri-can students. Having more parental support was associated with a morerapid decline in sense of belonging for members of both races, but thiseffect was more pronounced for white students. Finally, there was aninteraction between race and peer support on the rate of change ofsense of belonging (parameter estimate = )0.0028, p<0.05). For whitestudents, having more peer support was associated with a faster declinein sense of belonging over time. In contrast, for African American stu-dents having more peer support was associated with an increase in senseof belonging over time.Despite these few interactions with race, the results were remarkably

similar for members of both groups. It is of particular interest that theeffect of the intervention on sense of belonging, institutional commit-ment, or intentions to persist did not differ for white and AfricanAmerican students.

DISCUSSION

The main goal of this research was to highlight the importance ofsense of belonging in understanding student persistence, because, as


Hurtado and Carter (1997) have argued, it is a potentially importantbut seriously understudied variable. To do this, we examined the predic-tors of sense of belonging, including the effects of an interventiondesigned to enhance students’ sense of belonging. We also examinedwhether sense of belonging predicted institutional commitment andintentions to persist while controlling for student background variables(race, gender, financial difficulty, and SAT) and other variables thatcommonly predict persistence (peer interactions, faculty interactions,academic integration, peer support, and parental support).We found that several of these variables were associated with sense of

belonging at the beginning of the academic year. Students who reportedmore peer-group interactions, interactions with faculty, peer support,and parental support also initially reported having a greater sense ofbelonging. Neither academic integration nor student background vari-ables were associated with starting values of sense of belonging. It isinteresting to note that the variables that were associated with sense ofbelonging at the beginning of the year were all quite social in nature.That is, variables that pertained to interactions students had in the uni-versity setting (with peers or with faculty) or social support studentshad for entering the university setting (from peers or from parents) wereassociated with a greater sense of belonging, whereas students’ back-ground characteristics and academic integration were not. This suggeststhat the early social experiences students have when they first enter col-lege and the social support they receive during that time are likely to bebetter determinants of initial levels of sense of belonging than are demo-graphic characteristics or academic experiences.We also found that sense of belonging significantly declined over the

course of the academic year. Although academic integration was notassociated with sense of belonging at the start of the year (Time 1), itwas associated with sense of belonging’s rate of change. Having above-average academic integration was associated with an increase in sense ofbelonging over time, whereas having below-average academic integra-tion was associated with a decrease in sense of belonging over time.How well a student adjusts to the academic environment of college isthus closely tied to their developing sense of belonging with the college.This could be a fruitful area to target in programs to preserve students’sense of belonging over the course of their first year of college. Oncestudents begin college, taking measures to ensure that they become well-integrated academically may help guard against a decline in sense ofbelonging.Parental support was also a significant predictor of sense of belong-

ing’s rate of change, such that having more parental support was


associated with a more rapid decline in sense of belonging over time.Upon first glance, this effect seems counterintuitive. However, it is pos-sible that parents who detect their children’s waning sense of belongingto the university offer increased support in hopes of encouraging theirchildren to finish their degree. Alternatively, students who are more clo-sely connected to their parents could also be more tied to their homelives in general instead of to college, thus decreasing their sense ofbelonging with the university. Some evidence shows that students whogo home more often during college are more likely to drop out (Massey,Fischer, Lundry, and Charles, 2003).The intervention was successful in that it affected changes in students’

sense of belonging over time. Students who received mailings and giftsthat emphasized their status as a valued member of the university com-munity experienced a less rapid decline in their sense of belonging com-pared to students who received no gifts and to students in both controlgroups combined. Students who received gifts of any kind (i.e., the en-hanced sense of belonging and gift control groups combined) experi-enced a less rapid decline in sense of belonging than students whoreceived no gifts, but there was no difference in decline of sense ofbelonging for students in the gift control and no-gift control groups. Inaddition, those in the enhanced sense of belonging group experienced amarginally significant less rapid decline in sense of belonging over timethan those in the gift control group. A similar marginally significanteffect was found for intentions to persist. Thus, providing students withmailings and gifts that highlight their connection with the universitymay provide more protection against declining intentions to persist thanproviding comparable mailings that do not highlight the university con-nection.Although the intervention affected sense of belonging as it was

designed to do, the statistically significant effects of the intervention ten-ded to be small in size (see Table 6). The effects of this interventionremain important, however, because they demonstrate that students’sense of belonging over their first year of college can be affected by rela-tively simple and low cost means. More comprehensive and elaborateprograms targeting sense of belonging might well have a larger impactthan the intervention used in this study, which was designed merely totest the hypothesis that sense of belonging can be changed ratherthan as a full-fledged programmatic effort to foster students’ sense ofbelonging.In addition to being affected by the intervention, sense of belonging

was found to be a significant predictor of both institutional commitmentand intentions to persist, even after controlling for student background,


integration, and support variables. Sense of belonging was positivelyassociated with institutional commitment and intentions to persist at thestart of the academic year but was unrelated to changes over time ineither variable. Thus, even though sense of belonging is related to insti-tutional commitment and intentions to persist at the beginning of theyear, it appears that sense of belonging does not contribute to the devel-opment of commitment or intentions over the course of the first year ofcollege. This is somewhat surprising, given that it is reasonable toexpect students with a declining sense of belonging to become less com-mitted to their university and/or to display weaker intentions to finishtheir degree at their university. However, based on their early socialexperiences as they adjust to a new and sometimes daunting environ-ment, first year students may quite quickly develop a sense of whetherthey belong at their school, which then affects their early sense of insti-tutional commitment and intentions to persist. In contrast, as the schoolyear progresses more practical factors, such as the extent to which stu-dents view a college degree as a necessary means to desired ends (e.g.,entry into the career of their choice), may come to predominate indetermining their sense of commitment and their intentions to persist,accounting for the lack of impact of sense of belonging on change inthese variables.However, before placing too much emphasis on the lack of associa-

tion between sense of belonging and the developmental trajectories ofinstitutional commitment and intentions to persist, it is important torecognize that none of the variables studied were related to changes ininstitutional commitment. Furthermore, parental support was the onlystatistically significant predictor of changes in intentions to persist, asidefrom the marginally significant effect of the intervention. For institu-tional commitment, the UGM revealed that there was only marginallysignificant variability in rate of change over time, suggesting that thedecline in commitment is similar across students and making it unlikelythat any variables will be associated with rate of change in commitment.It is possible that if the study had followed students for a longer periodof time that more variation in students’ institutional commitment wouldhave emerged. The lack of variability observed during students’ firstyear could be because students enter college with very high levels ofcommitment, as our data suggest, and therefore all tend to experiencedecreases in that commitment over their first year as they react tocollege’s heavy academic demands. This tendency could overwhelm anypotential effects that individual differences in student experiences mighthave on commitment. Importantly this study’s results also suggest thatrelationships between commitment and peer-group interactions, peer


support, parental support, and sense of belonging exist at the beginningof the academic year rather than develop gradually over time.For intentions to persist, the UGM indicated that there was statisti-

cally significant variability in rate of change, making it more feasible forvariables associated with rate of change to be identified. However, thisstudy found parental support to be the only significant predictor ofchanges in intentions. Although there was a significant decline in inten-tions to persist over time, the absolute change over the course of thestudy was quite small given that most students still evidenced verystrong intentions to persist at the end of the academic year. Had thestudy followed students for a longer period of time, perhaps a moresubstantial decline in intentions would have emerged, making it morefeasible to identify significant predictors of such change.The positive relationship between institutional commitment and inten-

tions to persist was present at the start of the academic year, as was therelationship between sense of belonging and intentions to persist. Thissuggests that relationships among these variables may be present whenstudents enter college or develop very rapidly at the beginning of theschool year, but that later changes in intentions to persist are not neces-sarily the outcome of changes in commitment or sense of belonging.Although the main purpose of this research was to study sense of

belonging, a number of additional findings deserve note. Related to thepreceding discussion, this study found that several variables that are of-ten included in developmental models of student persistence displayedonly marginally significant variability in their rate of change duringfreshman year. Specifically, these variables were academic integration,peer support, parental support, and institutional commitment. This sug-gests that changes in these variables over time tend to be uniform,regardless of student backgrounds or college experiences. Researchersshould therefore be cautious when placing these variables within con-ceptual models that predict that these variables develop differently fordifferent students.It is also worth noting that several of the variables that exhibited a

significant change over time changed in the negative, rather than posi-tive, direction. Although peer-group interactions and interactions withfaculty increased over time, parental support, sense of belonging, insti-tutional commitment, and intentions to persist all declined over time.For parental support, perhaps students interact with their parents lessas they spend more time at college, so that the support their parents of-fer is less salient. It is also plausible that parents offer less support astheir children become more independent throughout the first year ofcollege. Either way, the decline in parental support is not surprising.


The general decline in sense of belonging, commitment, and intentionsto persist is most likely due to the excitement and high expectations stu-dents bring with them to college. As students become more familiarwith the challenging reality of being a college student, their initialexcitement may fade, resulting in a decline in sense of belonging, com-mitment, and intentions to persist.Another noteworthy finding of this research is that the effects of stu-

dent background variables (race, gender, SAT scores, and financial diffi-culty) were sparse in this sample. This might well occur in a universitythat draws a relatively homogeneous group of students who then reactand adjust to college in a similar manner. However, the students at theuniversity studied come from quite a range of academic and social back-grounds. Thus, it is possible that student background variables had rela-tively little impact on the variables included in this study because theuniversity at which the study was conducted does an effective job ofsocializing a diverse student body into a common university community.It certainly makes substantial efforts to do this. Whatever its cause, thelimited role of student background characteristics in the current data setis not consistent with the role of student background that has been pre-viously theorized. Specifically, Tinto (1993) hypothesized that studentbackground variables affect starting values of commitment and inten-tions to persist, which then affect integration experiences, etc. In thisresearch, the (few) significant effects of student background variableswere for the initial status of integration and support variables, ratherthan for sense of belonging, commitment, or intentions. Thus, the cur-rent findings do not support Tinto’s hypothesis that student backgroundvariables exert their influence on persistence decisions through theireffect on initial commitment and intentions.Finally, and related to the preceding discussion, it is important to

note that this study’s findings were remarkably similar for AfricanAmerican and white students. Of the many race interactions tested, onlya small number were significant. This is consistent with past researchthat has found models of student persistence to be quite similar forAfrican American and white students at predominantly-white universi-ties (e.g., Eimers and Pike, 1997; Mallinckrodt, 1988; Nettles, Thoeny,and Gosman, 1986). However, two differences between the effects ofparental and peer support on sense of belonging for African Americanand white students in this sample should be noted. Specifically, the posi-tive relationship between parental support and sense of belonging at thebeginning of the year was especially strong for African American stu-dents. Furthermore, although for white students peer support was asso-ciated with a faster decline in sense of belonging, for African American


students having more peer support was associated with an increase insense of belonging over time. These results suggest that peer and paren-tal support may be especially important factors for African Americanstudents in terms of the sense of belonging they develop in a predomi-nantly white university.

Limitations

One limitation of this study is its sample size, which is smaller thansamples used in other studies of student persistence (e.g., Biddle, Bank,and Slavings, 1987; Cabrera et al., 1993). The sample size was limitedby two factors. First, we attempted to recruit approximately the samenumber of white and African American freshmen, which meant that thesample size was constrained by the number of African American fresh-men at the university at the study’s beginning. Furthermore, because thestudy contained the intervention component and students were paid forcompleting the surveys, the sample size was constrained by the availabil-ity of financial resources and staff. A concerted effort was made toenroll eligible students into the study in order to maximize the samplesize. The type of statistical analysis utilized in this research does notrequire large samples, so the relatively small sample does not jeopardizethe validity of the analyses. However, as with any research that is basedon a sample of students at a single university, results should be general-ized to other institutions with caution.Another limitation is that this study focuses only on the first year of

college. The relationships among the variables reported here couldchange as students continue through college. However, the greatestamount of attrition occurs prior to the second year of college, therebymaking the freshman year an important time in which to study factorspredicting persistence (Barefoot, 2004). Furthermore, previous researchhas shown that models of student persistence during the freshman yearare similar to those for students in their later years of college (e.g.,Bean, 1985).

Conclusions

This research highlights the importance of including sense of belong-ing in empirical studies of student persistence. Sense of belonging wasfound to be significantly associated with institutional commitment at thestart of the academic year. Furthermore, even with student background,integration, and support variables in the model, sense of belonging andinstitutional commitment were the only two significant predictors of


intentions to persist at the start of the academic year. These findingssuggest that sense of belonging is an important but often overlookedvariable in studies of student persistence.This study also reports the effects of an intervention designed specifi-

cally to enhance students’ sense of belonging with the university. Itserves as evidence that the development of sense of belonging can bemodified using a relatively simple and inexpensive intervention. Giventhat sense of belonging was associated with the starting value, but notrate of change, of institutional commitment and intentions to persist,those who wish to develop future interventions using the current inter-vention as a model should consider starting the intervention prior to thestart of the academic year.Previous studies of persistence have conflated sense of belonging with

other constructs such as institutional fit or commitment (Bean, 1985;Cabrera et al., 1992, 1993; Nora and Cabrera, 1996; Strauss and Volkw-ein, 2004). Several aspects of the current findings suggest that sense ofbelonging should be added as a stand-alone variable in persistenceresearch. For instance, the developmental trajectories of sense ofbelonging and institutional commitment were associated with differentsets of predictors. The general decline in sense of belonging was attenu-ated by academic integration and magnified by parental support,whereas the decline in institutional commitment was the same for allstudents. Perhaps most important is the finding that the intervention,which was designed specifically to impact students’ sense of belonging,indeed protected against the decline in sense of belonging but had noeffect on changes in institutional commitment. In addition, sense ofbelonging predicted the initial status of intentions to persist, even aftercontrolling for institutional commitment. It thus appears that theunique role of sense of belonging in explaining student persistenceshould continue to be explored.Finally, this study is among the first to utilize multilevel modeling for

change analyses to study student persistence. As such, it provides ademonstration of another statistical tool that can be used to understandstudent persistence decisions. Importantly, its findings question existingmodels of student persistence that hypothesize that intentions to persistare the outcome of a developmental process that takes place throughoutthe first year of college. The results of this study suggest that many ofthe relationships between variables included in models of student persis-tence exist at or very near the beginning of students’ college careers andthat few of the commonly studied variables actually predict change overtime in things like institutional commitment and intentions to persist. Itis possible that student persistence decisions are the outcome of a


developmental process, but that this process takes place much earlierthan typically believed, such as between the time when students getaccepted into college and the first couple weeks of actually attendingclasses there. If this is the case, research should be conducted duringthis critical pre-college period and models of student persistence shouldtake into account student expectations of college rather than (or inaddition to) their actual experiences in college.

ACKNOWLEDGMENT

This research was funded by grants from the Spencer Foundation(Grant #200500074), the Sloan Foundation (Grant # B2004-46), theOffice of the Vice-Provost for Undergraduate Studies at the universityat which it was conducted, and by a Grants-in-Aid award to the firstauthor from the Society for the Psychological Study of Social Issues.We sincerely thank Marge Gibson, Jessica Girlardo, Amanda Good,Ashley Jones, Steve Luckey, Julie McDonald, Samantha McGinty,Wade Schaming, and Becki Shucosky for their assistance with thisresearch. We also thank Antonius Cillessen for sharing his data analyticwisdom and for offering helpful feedback on a previous draft of thispaper.

ENDNOTES

1. Details of the confirmatory factor analysis are available from the firstauthor upon request.

2. A third item from the Institutional and Goal Commitments subscalewas excluded from the analyses because it was conceptually redun-dant with the measure of intentions to persist and could thereforeartificially inflate the relationship between institutional commitmentand intentions to persist. This item asks students how likely it is thatthey will enroll at the university the following semester.

3. The goal commitment items from that subscale were not includedbecause others have found a lack of variance in goal commitmentamong some samples of college freshmen, making it difficult to usethis variable as a meaningful predictor in analyses (Berger andMilem, 1999; Milem and Berger, 1997). The alpha for Pascarella andTerenzini’s goal commitment items was also quite low in at least onepublished report (0.36, Mallette and Cabrera, 1991). Finally, whenseparate measures of institutional and goal commitment are included


in research, goal commitment tends to be less important as a predic-tor of student persistence (e.g., Bean, 1980; Cabrera et al., 1999;Eimers and Pike, 1997; Mallette and Cabrera, 1991).

4. Because the initial status intercepts merely indicate whether the aver-age starting value was different than the average over all time points,they will not be discussed further.

5. In MMC analyses, significant associations between time-varying pre-dictors (i.e., predictors that can have different values at different timepoints) and the initial status of an outcome represent associations be-tween the predictors measured at any time point and the initial status(i.e., Time 1 value) of the outcome. For example, the significant asso-ciation between the predictor variable peer-group interactions andthe initial status of the outcome variable sense of belonging meansthat students who reported having more peer-group interactions atthe beginning, middle, and/or end of their first year of college alsoreported a greater sense of belonging at the beginning of the aca-demic year.

6. The final contrast code (CC3B) is not of theoretical interest. None-theless, its results are reported in Table 5 for completeness. This con-trast was not significant for any outcome variable, indicating that theenhanced sense of belonging and no-gift control groups, combined,did not differ from the gift control group.

7. The full results of the models containing race interactions are avail-able from the first author upon request. Because there were so few,only significant race interactions are presented here.

REFERENCES

Astin, A. W. (1984). Student involvement: A developmental theory for higher education.

Journal of College Student Personnel 25(3): 297–308.

Bank, B. J., Slavings, R. L., and Biddle, B. J. (1990). Effects of peer, faculty and parental

influences on students’ persistence. Sociology of Education 63(3): 208–225.

Barefoot, B. O. (2004). Higher education’s revolving door: Confronting the problem of student

drop out in US colleges and universities. Open Learning 19(1): 9–18.

Baumeister, R. F., and Leary, M. R. (1995). The need to belong: Desire for interpersonal

attachments as a fundamental human motivation. Psychological Bulletin 117(3): 497–529.

Bean, J. P. (1980). Dropouts and turnover: The synthesis and test of a causal model of student

attrition. Research in Higher Education 12(2): 155–187.

Bean, J. P. (1985). Interaction effects based on class level in an explanatory model of college

student dropout syndrome. American Educational Research Journal 22(1): 35–64.


Berger, J. B., and Milem, J. F. (1999). The role of student involvement and perceptions of

integration in a causal model of student persistence. Research in Higher Education 40(6):

641–664.

Biddle, B. J., Bank, B. J., and Slavings, R. L. (1987). Norms, preferences, identities and

retention decisions. Social Psychology Quarterly 50(4): 322–337.

Bollen, K. A., and Hoyle, R. H. (1990). Perceived cohesion: A conceptual and empirical

examination. Social Forces 69(2): 479–504.

Cabrera, A. F., Castaneda, M. B., Nora, A., and Hengstler, D. (1992). The convergence

between two theories of college persistence. The Journal of Higher Education 63(2): 143–164.

Cabrera, A. F., Nora, A., and Castaneda, M. B. (1993). College persistence: Structural

equations modeling test of an integrated model of student retention. The Journal of Higher

Education 64(2): 123–139.

Cabrera, A. F., Nora, A., Terenzini, P. T., Pascarella, E., and Hagedorn, L. S. (1999). Campus

racial climate and the adjustment of students to college: A comparison between white

students and African-American students. The Journal of Higher Education 70(2): 134–160.

Cabrera, A. F., Stampen, J. O., and Hansen, W. L. (1990). Exploring the effects of ability to pay

on persistence in college. Review of Higher Education 13(3): 303–336.

Donovan, R (1984). Path analysis of a theoretical model of persistence in higher education

among low-income Black youth. Research in Higher Education 21(3): 243–259.

Durkheim, E. (1951). Suicide: A Study in Sociology (J. A. Spaulding & G. Simpson, Trans.).

Free Press, New York.

Eimers, M. T., and Pike, G. R. (1997). Minority and non-minority adjustment to college:

Differences or similarities? Research in Higher Education 38(1): 77–97.

Gaertner, S. L., and Dovidio, J. F. (2000). Reducing Intergroup Bias: The Common Ingroup

Identity Model, Psychology Press, Philadelphia, PA.

Hu, S., and St. John, E. P. (2001). Student persistence in a public higher education system:

Understanding racial and ethnic differences. The Journal of Higher Education 72(3):

265–286.

Hurtado, S. (1992). The campus racial climate: Contexts of conflict. The Journal of Higher

Education 63(5): 539–569.

Hurtado, S., and Carter, D. F. (1997). Effects of college transition and perceptions of the

campus racial climate on Latino college students’ sense of belonging. Sociology of Education

70(4): 324–345.

Judd, C. M., and McClelland, G. H. (1989). Data Analysis: A Model Comparison Approach,

Harcourt Brace Jovanovich, New York.

Luhtanen, R., and Crocker, J. (1992). A collective self-esteem scale: Self-evaluation of one’s

social identity. Personality and Social Psychology Bulletin 18(3): 302–318.

Mallette, B. I., and Cabrera, A. F. (1991). Determinants of withdrawal behavior: An

exploratory study. Research in Higher Education 32(2): 179–194.

Mallinckrodt, B (1988). Student retention, social support, and dropout intention: Comparison

of Black and White students. Journal of College Student Development 19(1): 60–64.

Massey, D. S., Fischer, C., Lundry, G., and Charles, C. Z. (2003). The Source of the River: The

Social Origins of Freshmen at America’s Selective Colleges and Universities, Princeton

University Press, Princeton, NJ.

Mayo, J. R., Murguia, E., and Padilla, R. V. (1995). Social integration and academic

performance among minority university students. Journal of College Student Development

36(6): 542–552.

Milem, J. F., and Berger, J. B. (1997). A modified model of college student persistence:

Exploring the relationship between Astin’s theory of involvement and Tinto’s theory of

student departure. Journal of College Student Development 38(4): 387–400.


Moores, K., and Klas, L. D. (1989). Comparing personal, social, and institutional variables for

university dropouts and those who persist. College Student Journal 23(1): 16–22.

Munro, B. H. (1981). Dropouts from higher education: Path analysis of a national sample.

American Educational Research Journal 18(2): 133–141.

Nettles, M. T., Thoeny, A. R., and Gosman, E. J. (1986). Comparative and predictive analyses

of black and white students’ college achievement and experiences. Journal of Higher

Education 57(3): 289–318.

Nora, A. (1987). Determinants of retention among Chicano college students: A structural

model. Research in Higher Education 26(1): 31–59.

Nora, A. (1990). Campus-based aid programs as determinants of retention among Hispanic

community college students. Journal of Higher Education 61(3): 312–331.

Nora, A., and Cabrera, A. F. (1993). The construct validity of institutional commitment:

A confirmatory factor analysis. Research in Higher Education 34(2): 243–262.

Nora, A., and Cabrera, A. F. (1996). The role of perceptions of prejudice and discrimination on

the adjustment of minority students to college. The Journal of Higher Education 67(2):

119–148.

Nora, A., Cabrera, A., Hagedorn, L. S., and Pascarella, E. (1996). Differential impacts of

academic and social experiences on college-related behavioral outcomes across different

ethnic and gender groups at four-year institutions. Research in Higher Education 37(4):

427–451.

Pascarella, E. T., Duby, P. B., and Iverson, B. K. (1983). A text and reconceptualization of a

theoretical model of college withdrawal in a commuter institution setting. Sociology of

Education 56(2): 88–100.

Pascarella, E. T., and Terenzini, P. T. (1980). Predicting freshman persistence and voluntary

dropout decisions from a theoretical model. The Journal of Higher Education 51(1): 60–75.

Pascarella, E. T., Terenzini, P. T., and Wolfle, L. M. (1986). Orientation to college and

freshman year persistence/withdrawal decisions. Journal of Higher Education 57(2): 155–175.

Singer, J. B., and Willett, J. B. (2003). Applied Longitudinal Data Analysis: Modeling Change

and Event Occurrence, Oxford University Press, New York.

Snider, K., and Dovidio, J. F. (1996). A Survey of the Racial Climate at Indiana State University,

Institutional Research and Testing, Indiana State University, Terre Haute, IN.

Spady, W. G. (1971). Dropouts from higher education: Toward an empirical model. Interchange

2: 38–62.

Strauss, L. C., and Volkwein, J. F. (2004). Predictors of student commitment at two-year and

four-year institutions. The Journal of Higher Education 75(2): 203–227.

Tinto, V. (1987). Leaving College: Rethinking the Causes and Cures of Student Attrition,

University of Chicago Press, Chicago.

Tinto, V. (1993). Leaving College: Rethinking the Causes and Cures of Student Attrition (2nd

edn.). University of Chicago Press, Chicago.

Zea, M. C., Reisen, C. A., Beil, C., and Caplan, R. D. (1997). Predicting intention to remain in

college among ethnic minority and nonminority students. The Journal of Social Psychology

137(2): 149–160.

Received March 15, 2006; accepted September 6, 2006


8. hausmann, leslie rm; schofield, janet ward; woods, rochelle l. sense of belonging as a predictor...

Documents

students sense of belonging

awhether sense of belonging

university of pittsburgh

role of students sense

year students

sense ofbelonging

predictors of persistence

university drive c